Arterial stiffness and wave reflections in renal transplant recipients

Transcription

1 Nephrol Dial Transplant (2007) 22: doi: /ndt/gfm379 Advance Access publication 27 June 2007 Original Article Arterial stiffness and wave reflections in renal transplant recipients Francis Verbeke 1, Wim Van Biesen 1, Patrick Peeters 1, Luc M. Van Bortel 2 and Raymond C. Vanholder 1 1 Ghent University Hospital, Nephrology section, Department of Internal Medicine and 2 Ghent University, Heymans Institute of Pharmacology, De Pintelaan 185, 9000 Gent, Belgium Abstract Background. Arterial stiffness predicts cardiovascular disease (CVD) events and has been well documented in haemodialysis patients. Information in renal transplant recipients (RTR), however, remains limited despite their higher CVD risk compared to the general population. We aimed to assess arterial stiffening and wave reflections in RTR and healthy controls and to evaluate which factors could explain potential differences. Methods. Carotid augmentation index (AI) and carotid-femoral pulse wave velocity (PWV) were measured in 200 RTR and 44 controls using applanation tonometry. The impact of traditional and nontraditional CVD risk factors was assessed using linear regression analysis. Glomerular filtration rate (GFR) was measured by 51 Cr-EDTA (RTR) and estimated using the abbreviated Modification of Diet in Renal Disease formula (RTR and controls). Results. After correction for age, blood pressure and anthropometry, AI and PWV remained % (P ¼ 0.04) and m/s (P ¼ 0.01) higher in RTR than controls, corresponding to a difference in vascular age of >10 years. In multivariate analysis, additional independent factors related to AI and PWV were GFR ( 1.8% and 0.19 m/s per 10 ml/min) and C-reactive protein (3.2% and 0.21 m/s per logarithm increase). Conclusions. Increased arterial stiffness and wave reflections in RTR are attributable to incomplete restoration of GFR and the presence of subclinical inflammation. Keywords: arterial stiffness; cardiovascular risk; inflammation; kidney function; kidney transplantation Correspondence and offprint requests to: Francis Verbeke, University Hospital Ghent, Department of Internal Medicine, Nephrology section, De Pintelaan 185 B-9000 Gent, Belgium. francis.verbeke@ugent.be 1 The authors wish it to be known that, in their opinion, F.V. is the first author to this work. Introduction Despite a favourable selection bias for patients undergoing kidney transplantation, and the partial removal of uraemia-related risk factors after transplantation [1], mortality from cardiovascular disease (CVD) still remains at least 3 5 times higher in renal transplant recipients (RTR) than in the general population [2]. As in other populations affected by chronic kidney disease (CKD), traditional risk factors do not fully explain the excess CVD risk in RTR [3], implying a role for less traditional risk factors. Accurate identification of the CVD risk in this population is warranted because of the longer potential lifetime exposure owing to the improved graft survival. Evaluation of intermediate endpoints such as markers of central arterial stiffening can be helpful for more accurate risk stratification at a stage when intervention may still modify this risk. Arterial stiffness is a pathophysiological cause of cardiac ischaemia as it adversely affects ventricular vascular coupling, resulting in a higher cardiac workload during systole and a decreased coronary oxygen supply during diastole measurement. Arterial stiffness is also an early marker of CVD, and parameters like aortic pulse wave velocity (PWV) and central augmentation index (AI) have been shown predictive of CVD and total mortality in several distinct populations [4 8]. This seemingly universal predictive value is in agreement with the general applicability of the underlying basic pathophysiological principles. PWV is considered the gold standard method for assessing central arterial stiffness [9]. AI represents the relative contribution of peripherally reflected waves to the pressure generated during ventricular ejection, which is dependent on PWV but also on timing, intensity and localization of wave reflections. AI and PWV can be obtained quickly using applanation tonometry, a non-invasive technique requiring a very limited degree of technical expertise, which makes these parameters suitable for use in daily clinical practice. ß The Author [2007]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 3022 F. Verbeke et al. The aim of the present study was (i) to evaluate indices of central arterial stiffening and wave reflection in RTR compared to healthy controls; (ii) to evaluate to which factors with possible pathophysiological impact these potential differences were related. Subjects and methods Subjects All patients transplanted for at least 3 months and in stable clinical condition were eligible for the study. They had to be free from acute infections, acute rejection or surgery within 1 month and from any recent (<3 months) cardiovascular event prior to inclusion. Combined organ transplants and patients under treatment for malignancy (except nonmetastatic cutaneous lesions) were excluded. Of the 205 screened patients, 203 agreed to participate but three were excluded because the investigations were of insufficient technical quality, leaving 200 patients for analysis. A group of 44 healthy persons served as controls. The study was approved by the Local Research Ethics Committee and written informed consent was obtained from each participant. Protocol for assessment of central PWV and AI Studies were conducted in a quiet environment after at least 10 min of supine rest. All measurements were performed by a single trained investigator (F.V.). Blood pressure was recorded in the dominant or non-fistula arm using a validated oscillometric device (Omron M4-I; Omron Corporation, Japan). Brachial artery (BA) waveforms were recorded with a high-fidelity micromanometer (SPC-301; Millar Instruments, TX, USA) and calibrated with the BA oscillometric systolic and diastolic blood pressure. Mean arterial pressure (MAP), derived from the time-integration of the BA waveform, and diastolic blood pressure were subsequently used to calibrate carotid artery waveforms as previously described [10]. The local carotid AI was calculated from pulse wave analysis (SphygmoCor; AtCor Medical, Sydney, Australia). Aortic PWV was measured using the same device by sequentially recording ECG-gated carotid and femoral artery waveforms. The path length was calculated by subtracting the distance between sternal notch and carotid recording site from the distance between sternal notch and femoral site. Intra- and intersession coefficients of variation obtained during reproducibility studies were 5.4% and 6.1% for AI, 2.9% and 5.2% for PWV, and 2.2% and 5.1% for MAP. Laboratory measurements Serum calcium, phosphorus, haematocrit, total cholesterol, HDL cholesterol, triglycerides and parathyroid hormone were determined using standard methodology in a single accredited laboratory. High sensitivity C-reactive protein (CRP) was measured using a particle-enhanced immunoturbidimetric assay and serum creatinine with a rate-blanked compensated Jaffé method, both on a Roche/Hitachi Modular P analyzer (Roche Diagnostics GmbH, Mannheim, Germany). All laboratory values from 1 month after transplantation until the study date, or from the last 12 months for patients transplanted >13 months, were averaged. For CRP and serum creatinine, a time-averaged mean was calculated in order to minimize the effect of shortterm elevations. The glomerular filtration rate (GFR) was estimated using the abbreviated Modification of Diet in Renal Disease (amdrd) formula and in addition was also measured directly by 51 Cr-ethylenediaminetetraacetic acid ( 51 Cr-EDTA) clearance in 151 patients. The measured GFR was determined from the elimination rate of the tracer measured in plasma samples 2, 3 and 4 h after intravenous injection of 3.7 MBq of 51 Cr-EDTA per 75 kg of body weight and absolute values were normalized for a body surface area of 1.73 m 2. Statistical analysis Data are expressed as means SD or medians with interquartile range depending on the distribution. The unpaired t-test and one-way ANOVA with planned contrasts, or the Wilcoxon rank sum test was used for comparing between group means of normally and not normally distributed variables, respectively. Differences in frequency were tested by 2 analysis. Adjustment for potential confounders when comparing patients and controls was achieved by matching as well as by entering these variables as predictors in a multiple linear regression model, with PWV and AI as dependent variables and a dummy variable for patients (RTR; coded ¼ 1) vs controls (CT; coded ¼ 0). CRP was log-transformed because of its skewed distribution. Within the group of RTR, stepwise multiple linear regression analysis was used to evaluate which factors were independently associated with PWV and AI. Predictor variables were selected based upon significant correlation in the present analysis or in previous studies, and included following relevant demographic, anthropometric, haemodynamic and biochemical variables: age, gender, length, weight, body mass index, smoking, diabetes, history of CVD, time since transplantation; haematocrit, serum levels of calcium, phosphorus, calcium-phosphorus product, parathyroid hormone, triglycerides, total cholesterol, HDL cholesterol; MAP and heart rate. Additionally, treatment with angiotensin antagonists, any antihypertensive drug, statins, antiplatelet agents, steroids and calcineurin inhibitors were included in the stepwise regression analysis to assess their potential confounding effect on the covariates of interest. Dummy variables were used for gender (women ¼ 0; men ¼ 1), current smoking and history of vascular disease (absent ¼ 0; present ¼ 1). Interactions known from literature were tested before analysing main effects. The possible non-linear relationship of PWV and AI with age was explored using a second order polynomial term for age. Model assumptions were verified by analysis of residuals. Data were analysed using SPSS software (version 12.0). All tests were two-sided and a P-value < 0.05 was considered significant. Results Demographic and haemodynamic parameters of patients and controls are shown in Table 1. Laboratory results of patients are shown in Table 2.

3 Arterial stiffness in renal transplant recipients 3023 Table 1. Characteristics of patients and controls Characteristic (units) Value a P Transplants (n ¼ 200) Controls (n ¼ 44) Age (years) <0.001 Length (cm) <0.001 Weight (kg) BMI (kg/m 2 ) Men (n) 111 (56) 28 (64) 0.4 Current smoker (n) 23 (12) 7 (16) 0.5 Diabetes (n) 31 (16) 0 History of CVD (n) 68 (34) 0 BA SBP (mmhg) b <0.001 BA DBP (mmhg) c <0.001 MAP (mmhg) <0.001 Heart rate (min 1 ) Serum creatinine (mg/dl) <0.001 amdrd (ml/min) <0.001 CRP (mg/l) 3.3 ( ) 0.9 ( ) <0.001 Pulse wave velocity (m/s) <0.001 Augmentation index (%) <0.001 a Values are means SD, medians (interquartile range) or numbers (%). b BA SBP, brachial systolic blood pressure. c BA DBP, brachial diastolic blood pressure. Table 2. Laboratory results of patients Test (units) Value a Haematocrit (%) Cr-EDTA clearance (ml/min) b Cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglycerides (mg/dl) 156 ( ) Calcium (mg/dl) Phosphorus (mg/dl) Ca P (mg 2 /dl 2 ) c Parathyroid hormone (pg/ml) 64 (45 111) a Values are means SD or medians (interquartile range). b Ca P, calcium-phosphorus product. c Subset of 151 patients. Single (1%), double (55%) or triple (44%) maintenance immunosuppressive regimens consisted of different combinations of steroids (65%), azathioprine (21%), mycophenolate mofetil (70%), cyclosporine A (49%), tacrolimus (22%) and/or sirolimus (17%). The majority (92%) of the patients was on antihypertensive treatment (median number of drugs ¼ 2), 46% received statins and 29% antiplatelet therapy. Patients had been transplanted for a median of 62.5 months (interquartile range: ) after a median time on dialysis of 27.2 months (interquartile range: ). Most transplantations (95%) were from cadaveric donors. RTR patients vs controls AI and PWV were higher in patients vs controls, whether adjustment for age, gender and MAP was performed by matching (Table 3) or by multivariate Table 3. Subset of patients with controls matched for age, gender and mean blood pressure Characteristic (units) Value a RTR subset (n ¼ 44) Controls (n ¼ 44) Age (years) Length (cm) BA SBP (mmhg) b BA DBP (mmhg) c MAP (mmhg) Heart rate (min 1 ) Pulse wave velocity (m/s) Augmentation index (%) a Values are means SD. b BA SBP, brachial systolic blood pressure. c BA DBP, brachial diastolic blood pressure. Table 4. Multiple linear regression models for augmentation index (AI) Model Coefficient SE b P Model 1. AI, controls and transplants: adjusted R 2 ¼ 0.59, P < a (Constant) RTR (vs CT) b Model 2. AI, transplants: adjusted R 2 ¼ 0.54, P < a (Constant) <0.001 amdrd (10 ml/min) Log (CRP) a Results adjusted for age, age 2, MAP, gender, diabetes, heart rate and length. b CT, healthy controls. analysis (Tables 4 and 5, model 1). Matching yielded an adjusted difference in AI of 10% which is comparable to the difference of 7.4% obtained by multivariate adjustment. For PWV, both approaches resulted in the same adjusted difference of 0.7 m/s. Multivariate analysis in RTR patients Age-related increases of PWV and AI were non-linear, indicated by a significant second order polynomial term for age, as found in healthy persons by McEniery et al. [11]. AI and PWV were negatively related to GFR and positively related to log (CRP) (Tables 4 and 5, model 2). These relationships remained significant after accounting for the effects of age, age 2, gender, diabetes, MAP, heart rate and length. The stepwise regression procedure eliminated the remaining predictor variables from the final AI and PWV models: time since transplantation, weight, body mass index, smoking, history of CVD; haematocrit, serum levels of calcium, phosphorus, calcium-phosphorus product, parathyroid hormone, triglycerides, total cholesterol, HDL cholesterol; and treatment with any antihypertensive drug, angiotensin antagonists, statins, antiplatelet agents, steroids and calcineurin inhibitors. P

4 3024 F. Verbeke et al. Table 5. Multiple linear regression models for pulse wave velocity (PWV) Model Coefficient SE b P Model 1. PWV, controls and transplants: adjusted R 2 ¼ 0.57, P < a (Constant) RTR (vs CT) b Model 2. PWV, transplants: adjusted R 2 ¼ 0.54, P < a (Constant) amdrd (10 ml/min) Log (CRP) a Results adjusted for age, age 2, MAP, gender, diabetes, heart rate and length. b CT, healthy controls. AI (%) PWV (m/s) A B amdrd<median amdrd<median amdrd>median amdrd>median CRP>median CRP<median CRP>median CRP<median Fig. 1. (A) AI according to CRP and estimated GFR. Data adjusted for age, MAP, length and heart rate. Patients with CRP > median and amdrd < median (black bar) had a higher AI (40%) compared to the other groups (P < 0.001). Patients with CRP<median and amdrd > median (white bar) had a lower AI (26%) compared to the other groups (P ¼ 0.03). (B) PWV according to CRP and estimated GFR. Data adjusted for age, MAP and gender. Patients with CRP > median and amdrd<median (black bar) had a higher PWV (8.8 m/s) compared to the other groups (P ¼ 0.04). Patients with CRP<median and amdrd > median (white bar) had a lower AI (26%) and PWV (7.9 m/s) compared to the other groups (P ¼ 0.005). Using 51 Cr-EDTA clearance instead of amdrd yielded comparable results. When dividing RTR into four subgroups according to CRP and GFR above and below the median, the group with the highest CRP and lowest GFR (Figure 1A and B, black bars) had a higher adjusted AI of 40% (P < 0.001) and PWV of 8.8 m/s (P ¼ 0.04) compared to the other groups (ANOVA with contrasts). Conversely, patients with a CRP below the median and an amdrd above the median (Figure 1A and B, white bars) had a lower AI of 26% (P ¼ 0.03) and PWV of 7.9 m/s (P ¼ 0.005) compared to the other groups (ANOVA with contrasts). Discussion The present study analysed parameters of arterial stiffness (PWV) and wave reflections (AI) in RTR using an established methodology, recently proposed as a standard method in a consensus statement by an expert panel [9]. We found that PWV and AI are higher in RTR compared to healthy persons, even after adjustment for confounding factors. Within the RTR population, GFR and levels of CRP were identified as the only additional independent factors related to these indices. These relationships underscore the importance of residual renal transplant function and subclinical inflammation as potential markers of and therapeutic targets in post-transplant CVD. Although CVD mortality in RTR patients is lower than in dialysis, it still remains substantially higher than in the global population. This difference is not fully explained by conventional CVD risk factors. Parameters of vascular stiffness and wave reflection more accurately predict CVD risk and have the potential of guiding and evaluating treatment decisions and effects. The degree of kidney dysfunction has been related to central PWV in patients with different stages of CKD [12], and an association between CRP and aortic PWV has been reported in patients with CKD stage 5 on haemodialysis [13] as well as in apparently healthy individuals [14]. This is in line with the recognition of decreased GFR and elevated levels of CRP as independent risk factors for CVD [15]. Reliable data on arterial stiffness/wave reflection in RTR using a validated methodology [9] are limited. Bahous et al. [16] reported a higher PWV in RTR than in healthy controls, as also observed in our study. In addition, they found a relationship with smoking and acute rejection, which in turn was related to chronic allograft nephropathy and doubling of serum creatinine. However, this study only included acceptors of living donors and no data on AI and CRP were available. Other studies reported peripheral but not central PWV [17,18], used indirect estimates [19], or evaluated transfer function-based AI [18,20], all of which give rise to biased results. In the present study, we measured central PWV directly and analysed carotid artery waveforms using the same [4,7,8,21,22], or a very similar [6,23] approach as in the landmark

5 Arterial stiffness in renal transplant recipients 3025 studies on hard endpoints. Recently, McEniery et al. [11] described in a large population of healthy persons a steeper increase of AI at young age and a more prominent change in PWV after the age of 50. We extended these observations to RTR, and confirmed quadratic age to be a predictor of PWV and AI. This means that if this non-linear relationship of AI and PWV with age is not taken into account, other linearly contributing markers can be missed. To the best of our knowledge, none of the studies on vascular stiffness in RTR has taken this into account. Finally, we did not rely on single-laboratory measurements as in other studies, but used arithmetic or time-averaged means of all values up to 12 months before the study, thus reducing error due to random fluctuations. In addition, no study evaluated the simultaneous effect of kidney function and CRP on PWV and AI in RTR. Both CRP/micro-inflammation and GFR are related to stiffness in the general population [4] and in different stages of CKD [12,13]. The present study demonstrates that also in RTR patients, CRP and GFR are related to PWV and AI. The impact of CRP and GFR and their joint effect is clinically relevant as can be appreciated from Figure 1. Compared to the group with the lowest values, patients with both a GFR below and a CRP above the median have a 14% higher AI and a 0.9 m/s higher PWV, an effect of similar magnitude as an increase in vascular age by at least 10 years [11]. In haemodialysis patients, each 1 m/s increase in PWV and each 10% increase in AI has been shown to be associated with a relative risk of death of 1.39 [4] and 1.51 [21], respectively. It is also important to note that 2/3 of the differences in PWV and AI between RTR and controls can be attributed to the lower GFR in RTR patients, with a 1.80% increase in AI (Table 4, model 2) and a 0.19 m/s increase in PWV (Table 5, model 2), per 10 ml/ min decrease in GFR. The remaining 1/3 difference is attributable to a difference in high sensitivity CRP: a 2.7-fold (1 natural logarithm) increase of CRP, e.g. from 0.9 to 2.4 mg/l, thus from normal to subclinical inflammation, corresponds to a 3.23% (Table 4, model 2) increase in AI and a 0.20 m/s increase in PWV (Table 5, model 2). Differences in CRP (1.3 log or 3.8-fold) and GFR (39 ml/min) between RTR and controls were larger (Table 1), and by multiplying these differences with the obtained regression coefficients, a large part of the difference in PWV and AI can be explained. However, renal transplant patients differ from healthy persons in many aspects, and the effect of other factors such as treatment effects and plasma levels of homocysteine and fetuin-a, have not been addressed in the present study. Therefore, it cannot be excluded that other factors, independently or by their relation with inflammation and/or decreased GFR, also contributed to the observed differences. Low-grade inflammation in RTR may represent a residual effect from before transplantation and/or be due to other sources than those encountered in the dialysis stage, such as chronic allograft nephropathy (CAN) [24], chronic infections such as periodontal disease [25], abdominal obesity [26] or type of immunosuppression (sirolimus) [27]. In view of the potential impact of subclinical inflammation, these sources should be searched for and re-mediated if present. Besides specific sources of inflammation, microinflammation may also be primarily an indicator of sustained arterial damage due to atherosclerosis. This is supported by the study of Varagunam et al. [28] where pre-transplantation CRP was an independent predictor of cardiovascular and all-cause mortality after transplantation. In parallel, pre-transplant CRP has also been shown to be predictive of CAN post-transplantation [29], again supporting the hypothesis that the process of atherosclerosis with the associated inflammation, which is already present before transplantation, is the primum movens of vascular damage after transplantation. CKD is another cardiovascular risk factor that gained widespread attention over the past few years, with epidemiological data pointing at an increased risk starting from a GFR below ml/min [30]. Notably, the majority of the renal transplant population should be considered to be affected by CKD [31]. Indeed, most patients in our study had CKD stages 3 5 (64%), while only 36% had a GFR 60 ml/min. Across this wide range of kidney function we detected an independent relationship between GFR and AI and PWV. Although microinflammation is a common feature of the uraemic state, GFR remained significant after accounting for CRP, which supports the idea that CKD is an independent risk factor for CVD. This CRP-independent relationship may indicate a potential role of uraemic toxins not directly involved in inflammation such as homocysteine, asymmetric dimethylarginine, inorganic phosphate, phenylacetic acid or dinucleoside polyphosphates [32]. Alternatively, decreased GFR may represent uraemia-related conditions that adversely affect arterial compliance such as medial calcification, insulin resistance [33], increased sympathetic nerve activity or activation of the local vascular rennin angiotensin system [34]. Surprisingly, no relationships between parameters of arterial stiffness/wave reflection and total cholesterol, HDL cholesterol or triglycerides were found. This could be due to confounding factors such as treatments increasing (calcineurin-inhibitors, rapamycin) as well reducing lipid levels (statins). A more likely explanation perhaps is that CRP and GFR are more potent predictors of CVD risk than serum cholesterol [35], as a recent study in the general population [14] also failed to detect a relationship between lipids and AI and PWV. Our study is a cross-sectional analysis of transplanted patients. Unfortunately, no data before renal transplantation on PWV or AI are available. In view of the long waiting times for cadaveric kidney transplantation, such an approach would be extremely impractical and would require a very long observation period. Our study was also an observational study, and no interventions to improve GFR or avoid subclinical

6 3026 F. Verbeke et al. inflammation were done, so that by no means can causal relations be drawn. Conclusions Central PWV and carotid AI are increased in RTR as compared to controls. These indices of arterial stiffness and wave reflection are independently related to CRP and transplant kidney function. Differences in GFR between RTR and controls explain 2/3 of the increase in PWV and AI, the remaining difference being attributable to chronic subclinical inflammation. Interventional trials will be required to determine causality and to evaluate the potential benefit of strategies that reduce inflammation, protect against loss of kidney function and/or improve vessel wall elasticity. Alternatively, chronic elevations of CRP and poor transplant GFR may also identify kidney transplant recipients at high risk for CVD and prompt a more intensive control of other risk factors. Conflict of interest statement. None declared. References 1. Wolfe RA, Ashby VB, Milford EL et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999; 341: Sarnak MJ, Levey AS, Schoolwerth AC et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003; 108: Kasiske BL, Chakkera HA, Roel J. Explained and unexplained ischemic heart disease risk after renal transplantation. J Am Soc Nephrol 2000; 11: Blacher J, Guerin AP, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness on survival in end-stage renal disease. Circulation 1999; 99: Meaume S, Benetos A, Henry OF, Rudnichi A, Safar ME. Aortic pulse wave velocity predicts cardiovascular mortality in subjects >70 years of age. Arterioscler Thromb Vasc Biol 2001; 21: Cruickshank K, Riste L, Anderson SG, Wright JS, Dunn G, Gosling RG. Aortic pulse-wave velocity and its relationship to mortality in diabetes and glucose intolerance: an integrated index of vascular function? Circulation 2002; 106: Laurent S, Boutouyrie P, Asmar R et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension 2001; 37: Laurent S, Katsahian S, Fassot C et al. Aortic stiffness is an independent predictor of fatal stroke in essential hypertension. Stroke 2003; 34: Laurent S, Cockcroft J, Van Bortel L et al. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27: Verbeke F, Segers P, Heireman S, Vanholder R, Verdonck P, Van Bortel LM. Non-invasive assessment of local pulse pressure: importance of brachial-to-radial pressure amplification. Hypertension 2005; 46: McEniery CM, Yasmin, Hall IR, Qasem A, Wilkinson IB, Cockcroft JR. Normal vascular aging: differential effects on wave reflection and aortic pulse wave velocity: the Anglo-Cardiff Collaborative Trial (ACCT). J Am Coll Cardiol 2005; 46: Wang MC, Tsai WC, Chen JY, Huang JJ. Stepwise increase in arterial stiffness corresponding with the stages of chronic kidney disease. Am J Kidney Dis 2005; 45: London GM, Marchais SJ, Guerin AP, Metivier F, Adda H, Pannier B. Inflammation, arteriosclerosis, and cardiovascular therapy in hemodialysis patients. Kidney Int 2003; 63 [Suppl. 84]: S88 S Yasmin, McEniery CM, Wallace S, Mackenzie IS, Cockcroft JR, Wilkinson IB. C-reactive protein is associated with arterial stiffness in apparently healthy individuals. Arterioscler Thromb Vasc Biol 2004; 24: European Society of Hypertension-European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens 2003; 21: Bahous SA, Stephan A, Blacher J, Safar ME. Aortic stiffness, living donors, and renal transplantation. Hypertension 2006; 47: Zoungas S, Kerr PG, Chadban S et al. Arterial function after successful renal transplantation. Kidney Int 2004; 65: Covic A, Goldsmith DJ, Gusbeth-Tatomir P, Buhaescu I, Covic M. Successful renal transplantation decreases aortic stiffness and increases vascular reactivity in dialysis patients. Transplantation 2003; 76: Kneifel M, Scholze A, Burkert A et al. Impaired Renal Allograft Function is Associated with Increased Arterial Stiffness in Renal Transplant Recipients. Am J Transplant 2006; 6: Ferro CJ, Savage T, Pinder SJ, Tomson CR. Central aortic pressure augmentation in stable renal transplant recipients. Kidney Int 2002; 62: London GM, Blacher J, Pannier B, Guerin AP, Marchais SJ, Safar ME. Arterial wave reflections and survival in end-stage renal failure. Hypertension 2001; 38: Willum-Hansen T, Staessen J, Torp-Pedersen C et al. Prognostic value of aortic pulse wave velocity as index of arterial stiffness in the general population. Circulation 2006; 113: Sutton-Tyrrell K, Najjar S, Boudreau R et al. Health ABC Study. Elevated aortic pulse wave velocity, a marker of arterial stiffness, predicts cardiovascular events in well-functioning older adults. Circulation 2005; 111: Sezer S, Akcay A, Ozdemir FN, Kulah E, Arat Z, Haberal M. Post-transplant C-reactive protein monitoring can predict chronic allograft nephropathy. Clin Transplant 2004; 18: Franek E, Blach A, Witula A et al. Association between chronic periodontal disease and left ventricular hypertrophy in kidney transplant recipients. Transplantation 2005; 80: van Ree RM, de Vries AP, Oterdoom LH et al. Abdominal obesity and smoking are important determinants of C-reactive protein in renal transplant recipients. Nephrol Dial Transplant 2005; 20: Thaunat O, Beaumont C, Chatenoud L et al. Anemia after late introduction of sirolimus may correlate with biochemical evidence of a chronic inflammatory state. Transplantation 2005; 80: Varagunam M, Finney H, Trevitt R et al. Pretransplantation levels of C-reactive protein predict all-cause and cardiovascular mortality, but not graft outcome, in kidney transplant recipients. Am J Kidney Dis 2004; 43: Fink JC, Onuigbo MA, Blahut SA et al. Pretransplant serum C-reactive protein and the risk of chronic allograft nephropathy in renal transplant recipients: a pilot case-control study. Am J Kidney Dis 2002; 39: Vanholder R, Massy Z, Argiles A, Spasovski G, Verbeke F, Lameire N. Chronic kidney disease as cause of cardiovascular morbidity and mortality. Nephrol Dial Transplant 2005; 20: Marcen R, Pascual J, Tenorio M et al. Chronic kidney disease in renal transplant recipients. Transplant Proc 2005; 37:

7 Arterial stiffness in renal transplant recipients Vanholder R, Glorieux G, Lameire N. New insights in uremic toxicity. Contrib Nephrol 2005; 149: Shinohara K, Shoji T, Tsujimoto Y et al. Arterial stiffness in predialysis patients with uremia. Kidney Int 2004; 65: Amann K, Neususs R, Ritz E, Irzyniec T, Wiest G, Mall G. Changes of vascular architecture-independent of blood-pressure in experimental uremia. Am J Hypertens 1995; 8: Blake GJ, Ridker PM. Novel clinical markers of vascular wall inflammation. Circ Res 2001; 89: Received for publication: Accepted in revised form: